<p>No way is UVA tier 2 with those other schools. You also have NYU too high. This is going by what Schmaltz’ parameters are for school comparisons.</p>
<p>No. Tier 1 is current top 50. Tier 2 is current 51-100. Tier 3 is 101-150 etc. Then list them alphabetically. No more hairsplitting.</p>
<p>
</p>
<p>Fixed for ya, rjk. ;)</p>
<p>Alam, call them whatever you want. That type of cosmetic difference has no impact. OK to call them “groups”?</p>
<p>Novi, the crux of my approach is not which schools go where. The crux is that schools in adjacent “groups” are considered comparable in terms of quality and prestige. I allows people to comprehend that while Group 1 schools are comparable to group 2 schools, and group 2 schools are comparable to groups 1 and 3, Group 1 and 3 AREN’T comparable. So if a student is trying to decide between a group 1 school and a group 2 school, he would be told that there is no significant difference in quality or prestige. Likewise if he were considering schools in group 2 and 3. But if the 2 schools he’s considering are TWO or more groups apart (e.g., in groups 1 and 3), THEN there would be a significant difference. We’re used to thinking that if A=B and B=C, then A=C. But I really don’t think that’s how colleges work.</p>
<p>Ghost, make the “groups” however big you want them to be, but in order for this system to make sense, adjacent groups need to be quite comparable in quality. So in your view, 1-100 are all comparable?</p>
<p>It seems like everybody is focusing on the groups, but the key here is that the groups are imprecise enough that colleges need to be TWO groups apart (not just 1 group apart) in order for there to be a significant difference.</p>
<p>For undergraduate schools, why not a two tier system?:</p>
<p>Tier 1
Princeton</p>
<p>Tier 2
all others</p>
<p>hawkette
but then again, 99% of Cal’s students were from the top 10% of their HS class and Cal does not weigh SATs as they do HS gpa, and does not superscore, whilst almost all privates do. I wonder how are you going to reconcile that…</p>
<p>JohnAdams, I agree wholeheartedly.</p>
<p>
</p>
<p>Really, all we need are those two statements.</p>
<p>1) Your system is essentially the same as everyone else’s, except that adjacent tiers are comparable while others are not (note that the placement of colleges into tiers is totally arbitrary)
2) Your justification for the correctness of this system is that it’s right because you said so.</p>
<p>I appreciate the effort, but I think the product is total fail.</p>
<p>People on this site swoon for Princeton because it’s said to strike the right balance of “undergrad focus” and top academic programs across a broad range of disciplines (that are mainly derived from having great profs and graduate schools).</p>
<p>…But even Princeton ain’t for everyone.</p>
<p>“Novi, the crux of my approach is not which schools go where. The crux is that schools in adjacent “groups” are considered comparable in terms of quality and prestige.”</p>
<p>Precisely why UVA was rated too high by the image. Virginia is not one tier below HYPSMC. It is no better than Michigan or UCLA.</p>
<p>RML,
Your desperation to promote UC Berkeley is such fun because you so grossly overrate its place among American undergraduate colleges. </p>
<p>As a foreigner, maybe you don’t realize it, but for making comparisons of top American colleges, using class rank is an absolute joke. There are lots of places with a higher quality student body than UC Berkeley, including that little place across the Bay which only has a pitiable 92% Top 10% students. </p>
<p>Furthermore, using your criteria, a half dozen UCs would be superior to Stanford (UCLA at 97%, UCSD at 100%, Davis at 98%, Santa Barbara at 96%, Irvine at 96%, Santa Cruz at 96%, even Riverside at 94%). LOL. </p>
<p>Btw, do you realize that only 67% of UC Berkeley students have a verifiable Top 10% rank? </p>
<p>4261 = # of UC Berkeley entering freshmen, Fall 2009 (99% Top 10%)
4225 = # of UC Berkeley entering freshmen, Fall 2008 (99% Top 10%)
4138 = # of UC Berkeley entering freshmen, Fall 2007 (99% Top 10%)
4059 = # of UC Berkeley entering freshmen, Fall 2006 (99% Top 10%)</p>
<p>2012 = # of UC Berkeley transfers, Fall 2009 (class rank data not verifiable)
2036 = # of UC Berkeley transfers, Fall 2008 (class rank data not verifiable)
1971 = # of UC Berkeley transfers, Fall 2007 (class rank data not verifiable)
1994 = # of UC Berkeley transfers, Fall 2006 (class rank data not verifiable)</p>
<p>24,529 = the 4-year total number of undergraduate students </p>
<p>16,516, 67%= the number and percentage of students with a verifiable Top 10% rank</p>
<p>I wonder how you are going to reconcile that…</p>
<p>Also, one of the dirty little secrets of the CDS is that the SAT data is superscored. Talk to a candid public school Adcomm and you learn the truth about what is going on. Even if the admissions process is done on a single sitting basis, the reported CDS data is superscored. Also, if you dispute this, compare the reported ACT data and you will see confirmation of this.</p>
<p>I hate the Berkeley/UMich prestige battles on these forums. Ugh.</p>
<p>At least USNews is nice enough to make their methodology available to criticize…</p>
<p>Why is Caltech always included on these lists? Yes, it is outstanding for what it offers, but it is a specialty school with what, 900 students? Caltech cannot possibly be considered comparable in “quality” to all of those other top tier schools - it does not even offer an English major. 99.99% of high school students in America have no absolutely no interest in attending Caltech.</p>
<p>
hawkette, your desperation to promote regional private research universities is such fun because you so grossly overrate their place among American research universities. ;)</p>
<p>Is the main point here simply that minor ranking differences should be considered insignificant in choosing a college? If so, I don’t understand how this could be at all controversial. What seems to be controversial is how we group schools. Are 5 positions in a magazine ranking significant? 10? 25? What principled basis is there for deciding?</p>
<p>There may be only the most minute quality differences between #1 and #10, but relatively large differences between #10 and #30, in some of the 4 factors hawkette cites. So if you plot these differences, maybe you’d get a sharp bend in the curve after school number N, or cluster M, for one quality factor. </p>
<p>For example, about 70-75% of classes at most of the top 10 or 15 national universities have fewer than 20 students. You get many small classes at virtually all these schools. Below ranking #30 or so (and among public universities), that percentage drops below 50% at many schools. If you highly value small discussion classes, this may be a sensible way to differentiate 2 or more groups. But hawkette’s other 3 quality factors may or may not track to the same line. So it’s not clear to me that quality clusters (“tiers”) do, or don’t, have any empirical basis when you aggregate multiple factors.</p>
<p>The clustering choices that do have a clear empirical basis for many students are “fit” decisions (for cost, location, size, climate, public v. private, coed v. women’s, etc.) If you can decide that you want a small, selective, private liberal arts college in the upper Midwest, you can quickly and clearly identify a small set of target schools.</p>
<p>
I’m very much looking forward to the potential Duke-Berkeley NCAA game. :D</p>
<p>^ I’m not counting my chickens before they hatch.</p>
<p>Baelor, I’m guessing most people would agree that regardless of what ranking is used (US News or whatever), schools within a few spots of each other are comparable (that is, college rankings aren’t precise enough that a difference of a few spots is significant).</p>
<p>But if you believe that with colleges a few spots apart A=B, and B=C, therefore A=C, there is nothing to stop you from carrying out the pattern until you get to A=Z and beyond. If that’s fine with you, then eventually you’ll get to Princeton = Southwest Missouri Barber College.</p>
<p>If it’s not ok with you, and you still think minor differences in rankings are insignificant (and therefore it’s a problem that the weakest school in group 1 is considered “better” than the best school in group 2), you need a method that will allow you to roughly equate group 1 and group 2 schools, without allowing the chain to continue infinitely to A=Z and beyond. That’s where the requirement that the comparable groups be only one group apart comes in.</p>
<p>I am not disagreeing with you on principal.</p>
<p>I am just saying that your system is inherently just as worthless as everyone else’s because it’s totally subjective. You are arbitrarily assigning tiers to universities, and moreover arbitrarily deciding how many tiers are necessary to separate schools substantially.</p>
<p>How again is this a resolution?</p>
<p>^ Exactly. The idea of using tiers is not a bad one, but your particular tiers are useless.</p>